The implementation of systematic testing for hepatitis B virus (HBV) has been instrumental in eliminating this virus from the blood supply. Nevertheless, a significant number of post-transfusion hepatitis (PTH) cases still occur. These cases are generally attributable to non-A, non-B hepatitis (NANBH) virus(es), the diagnosis of which is usually made by exclusion of other viral markers.
The etiological agent responsible for a large proportion of these cases has recently been cloned (Choo, Q-L et al. Science (1988) 244:359-362) is and a first-generation antibody test developed (Kuo, G. et al. Science (1989) 244:362-364). The agent has been identified as a positive-stranded RNA virus, and the sequence of its genome has been partially determined. Studies suggest that this virus, referred to subsequently as hepatitis C virus (HCV), may be related to flaviviruses and pestiviruses. A portion of the genome of an HCV isolated from a chimpanzee (HCVCDC/CHI) is disclosed in EPO 88310922.5. The coding sequences disclosed in this document do not include sequences originating from the 5xe2x80x2-end of the viral genome which code for putative structural proteins. Recently however, sequences derived from this region of the HCV genome have been published (Okamoto, H. et al., Japan J. Exp. Med. 60:167-177, 1990.). The amino acid sequences encoded by the Japanese clone HC-J1 were combined with the HCVCDC/CHI sequences in a region where the two sequences overlap to generate the composite sequence depicted in FIG. 1. Specifically, the two sequences were joined at glycine451. It should be emphasized that the numbering system used for the HCV amino acid sequence is not intended to be absolute since the existence of variant HCV strains harboring deletions or insertions is highly probable. Sequences corresponding to the 5xe2x80x2 end of the HCV genome have also recently been disclosed in EPO 90302866.0.
In order to detect potential carriers of HCV, it is necessary to have access to large amounts of viral proteins. In the case of HCV, there is currently no known method for culturing the virus, which precludes the use of virus-infected cultures as a source of viral antigens. The current first-generation antibody test makes use of a fusion protein containing a sequence of 363 amino acids encoded by the HCV genome. It was found that antibodies to this protein could be detected in 75 to 85% of chronic NANBH patients. In contrast, only approximately 15% of those patients who were in the acute phase of the disease, had antibodies which recognized this fusion protein (Kuo, G. et al. Science (1989) 244:362-364). The absence of suitable confirmatory tests, however, makes it difficult to verify these statistics. The seeming similarity between the HCV genome and that of flaviviruses makes it possible to predict the location of epitopes which are likely to be of diagnostic value. An analysis of the HCV genome reveals the presence of a continuous long open reading frame. Viral RNA is presumably translated into a long polyprotein which is subsequently cleaved by cellular and/or viral proteases. By analogy with, for example, Dengue virus, the viral structural proteins are presumed to be derived from the amino-terminal third of the viral polyprotein. At the present time, the precise sites at which the polyprotein is cleaved can only be surmised. Nevertheless, the structural proteins are likely to contain epitopes which would be useful for diagnostic purposes, both for the detection of antibodies as well as for raising antibodies which could subsequently be used for the detection of viral antigens. Similarly, domains of nonstructural proteins are also expected to contain epitopes of diagnostic value, even though these proteins are not found as structural components of virus particles.